Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages

Abstract

Object detection has made great progress in the past few years along with the development of deep learning. However, most current object detection methods are resource hungry, which hinders their wide deployment to many resource restricted usages such as usages on always-on devices, battery-powered low-end devices, etc. This paper considers the resource and accuracy trade-off for resource-restricted usages during designing the whole object detection framework. Based on the deeply supervised object detection (DSOD) framework, we propose Tiny-DSOD dedicating to resource-restricted usages. Tiny-DSOD introduces two innovative and ultra-efficient architecture blocks: depthwise dense block (DDB) based backbone and depthwise feature-pyramid-network (D-FPN) based front-end. We conduct extensive experiments on three famous benchmarks (PASCAL VOC 2007, KITTI, and COCO), and compare Tiny-DSOD to the state-of-the-art ultra-efficient object detection solutions such as Tiny-YOLO, MobileNet-SSD (v1 & v2), SqueezeDet, Pelee, etc. Results show that Tiny-DSOD outperforms these solutions in all the three metrics (parameter-size, FLOPs, accuracy) in each comparison. For instance, Tiny-DSOD achieves 72.1% mAP with only 0.95M parameters and 1.06B FLOPs, which is by far the state-of-the-arts result with such a low resource requirement.

arXiv comment12 pages, 3 figures, accepted by BMVC 2018

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• Caffe: Convolutional Architecture for Fast Feature Embedding
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• DSSD : Deconvolutional Single Shot Detector
• ParseNet: Looking Wider to See Better
• YOLO9000: Better, Faster, Stronger
• Xception: Deep Learning with Depthwise Separable Convolutions
• DSOD: Learning Deeply Supervised Object Detectors from Scratch
• Subcategory-aware Convolutional Neural Networks for Object Proposals and Detection

Cited by in corpus (20)

• A Survey of Deep Learning-based Object Detection
• Object Detection in 20 Years: A Survey
• A Smart, Efficient, and Reliable Parking Surveillance System with Edge Artificial Intelligence on IoT Devices
• Towards Accurate One-Stage Object Detection with AP-Loss
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• ATRW: A Benchmark for Amur Tiger Re-identification in the Wild
• AP-Loss for Accurate One-Stage Object Detection
• Cluster Pruning: An Efficient Filter Pruning Method for Edge AI Vision Applications
• ThunderNet: Towards Real-time Generic Object Detection
• EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse
• RefineDetLite: A Lightweight One-stage Object Detection Framework for CPU-only Devices
• YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design
• Recent Advances in Deep Learning for Object Detection
• Rethinking Channel Dimensions for Efficient Model Design
• CircleNet: Reciprocating Feature Adaptation for Robust Pedestrian Detection
• Efficient Deep Neural Networks
• Sem-LSD: A Learning-based Semantic Line Segment Detector
• FrostNet: Towards Quantization-Aware Network Architecture Search
• Understanding the Effects of Pre-Training for Object Detectors via Eigenspectrum
• Object Detection from Scratch with Deep Supervision